Chronic disease dietary therapy intelligent optimization method for non-nutrient regulation through multiple key targets

Abstract

The invention provides a chronic disease diet therapy intelligent optimization method for performing non-nutrient regulation through multiple key targets, and in combination with non-nutrient influence key targets influenced by chronic disease key targets, a non-nutrient-based optimized diet prescription is recommended to chronic disease patients with corresponding symptoms, so that the oxidation resistance, inflammation resistance and metabolism regulation of the chronic disease patients are improved. Through the influence of diet mode intervention based on the family nurse diet therapy theory on the metabolic index of the chronic disease common basic metabolic syndrome patient, the actual effect of a previous diet therapy scheme is optimized, and a diet mode capable of being popularized is formed. According to the method, the intelligent algorithm model is combined to carry out chronic disease diet therapy optimization intervention, the blank that a healthy diet mode and a related guide disease treatment principle are not clear is filled based on the basic principle of non-nutrient chronic disease intervention, and diet therapy tracking management is carried out on chronic disease patients through the system; and a chronic disease diet therapy full-optimal recipe corresponding to the real-time condition of the patient is dynamically adjusted, so that the rehabilitation and rehabilitation of the chronic disease of the patient are promoted.

Description

Technical Field

[0001] The present invention relates to the field of intelligent life management technology, and in particular to an intelligent optimization method for chronic disease diet therapy that uses multiple key targets to perform non-nutrient regulation, as well as an intelligent optimization device, electronic device, and computer-readable storage medium for chronic disease diet therapy that uses multiple key targets to perform non-nutrient regulation. Background Art

[0002] According to the classification standards proposed by the World Health Organization / Food and Agriculture Organization of the United Nations / International Federation of Health, Sports, and Physical Education, nutrients include nine amino acids in protein, two polyunsaturated fatty acids in lipids, one carbohydrate, seven macronutrients, eight trace elements, 14 vitamins, and water, for a total of 42 nutrients. Non-nutrients do not provide nutrients, but they help the body digest and absorb nutrients, promote metabolism, and maintain human health. Non-nutrients are active ingredients found in botanical herbs and foods that have a chemical structure different from that of nutrients and are dissolved in media such as water or ethanol. A growing number of studies have shown that various non-nutrient components in the diet play an indispensable role in maintaining human physiological functions and preventing certain diseases.

[0003] In 1975, Jenkins et al. published the first report in the Lancet confirming the regulatory effects of pectin, colloids, and wheat fiber on serum cholesterol, which garnered widespread attention from the scientific community. While non-nutrients in food are relatively few, they are diverse and complex, and currently lack a clear classification. Non-nutrients specifically refer to plant-based bioactive substances and dietary fiber that are common components of a healthy dietary pattern, excluding probiotics and other food additives. The range of phytochemicals within non-nutrients is vast, and a diet high in fruits, vegetables, and grains can effectively prevent chronic diseases. Plant-based bioactive compounds are numerous, including flavonoids, phytoestrogens, phenolic acid compounds, carotenoids, and sulfides. In Mediterranean diets, there are at least 5,000 individual phytochemicals found in fruits, vegetables, and grains, and many more remain unknown.

[0004] Non-nutrients can intervene in chronic diseases. Polyphenols are a major component of non-nutrients. Phenolic compounds in edible plants generally include phenolic acids, flavonoids, lignans, coumarins, and tannins. Flavonoids are the most widely distributed in plants, accounting for two-thirds of dietary polyphenols. The remaining one-third is phenolic acids, making them the second most important phenolic group in food. Phenolic compounds in plant foods that are relevant to human health include soy isoflavones, catechins, quercetin, anthocyanins, proanthocyanidins, curcumin, and resveratrol. Flavonoids typically exist as glycosides. Phenolic compounds are present in plants and, during various stages of processing, undergo a series of reactions to form complex phenolic compounds. Research has shown that phenolic compounds have antioxidant and anti-infective properties, and can reduce the risk of cardiovascular disease. They also possess certain pharmacological properties, such as anti-infective, anti-viral, anti-bacterial, anti-allergic, anti-hemorrhagic, and immune-enhancing properties. Phenolic compounds are widely distributed plant secondary metabolites in nature and are abundant in fruits and vegetables.

[0005] Domestic dietary interventions for chronic diseases are primarily designed based on the specific characteristics of each chronic disease. For example, dietary treatments for hypertension include limiting sodium intake, increasing potassium intake, controlling weight, limiting alcohol consumption, and increasing dietary fiber intake. Recommended foods include celery, onions, garlic, and kelp. For diabetes, dietary treatment principles include controlling total calories, rationing carbohydrates, consuming moderate amounts of protein, controlling fat intake, consuming a high-fiber diet, and eating small, frequent meals. Recommended foods include bitter melon, pumpkin, oats, and Chinese yam. However, these approaches do not provide dietary plans that optimize non-nutrient diets for chronic disease patients (primarily providing nutrient diets). They also fail to optimize the effectiveness of existing dietary interventions and develop scalable dietary patterns through dietary interventions based on family nurse dietary therapy theory (based on non-nutrient chronic disease diets) and by studying the effects on metabolic markers of patients with metabolic syndrome, a common underlying chronic disease. Consequently, existing dietary interventions for chronic diseases are relatively simplistic and lack comprehensive management of non-nutrient chronic disease diets. Consequently, they fail to comprehensively consider key influencing factors in chronic disease patients to optimize and recommend non-nutrient diets, resulting in a lack of intelligent management.

[0006] In addition, the existing dietary intervention methods for chronic diseases have not formed a closed-loop management system for the intervention of chronic disease patients. Therefore, it is impossible to form an effective plan for dietary intervention for chronic disease patients, and it is not practical and popularizable enough. Summary of the Invention

[0007] In order to solve the technical problems existing in the prior art, the present invention provides the following technical solutions:

[0008] On the one hand, a method for intelligent optimization of chronic disease diet therapy by regulating non-nutrients through multiple key targets is provided, which is implemented by an electronic device and includes:

[0009] S1. Based on the classification algorithm model, build and deploy a chronic disease diet recommendation model for non-nutrients based on key chronic disease targets and corresponding chronic disease conditions;

[0010] S2. Obtaining in real time the chronic disease status of chronic disease patients and examination data under various chronic disease assessment indicators from the chronic disease electronic medical records in the HIS system, wherein the chronic disease assessment indicators include morphological indicators and physiological and biochemical indicators. The morphological indicators include waist circumference, blood pressure, height, weight or BMI. The physiological and biochemical indicators include the following key target indicators of chronic diseases: oxidative stress, inflammation and metabolic disorders;

[0011] S3. Importing the chronic disease status of the chronic disease patient and the examination data under various chronic disease assessment indicators into the chronic disease diet therapy recommendation model, and having the chronic disease diet therapy recommendation model recommend a corresponding non-nutritional vegetarian diet therapy list;

[0012] S4. Extract keywords from the examination data under the chronic disease status and each chronic disease assessment indicator, and combine them with the pre-built diet recommendation logic to generate corresponding prompt words and input them into the preset LLM large language model. Use the LLM large language model to search the chronic disease recipe intervention database to retrieve a recommended diet therapy list that matches the prompt words;

[0013] S5. Optimize the recommended dietary therapy list using the non-nutritional vegetarian therapy list, replace or supplement the non-nutrients therein, obtain a complete and optimal diet recipe for chronic diseases, and recommend it to the user terminal;

[0014] S6. Update the electronic medical record of the chronic disease according to the diet therapy cycle, and repeat the above steps S2-S5.

[0015] Optionally, S1. Based on the classification algorithm model, a chronic disease diet recommendation model is constructed and deployed for recommending non-nutrients for corresponding chronic disease conditions based on key targets of chronic diseases, including:

[0016] Collect a structured dataset M consisting of non-nutrients that intervene in the key targets of chronic diseases under different chronic disease conditions and the key targets of chronic diseases:

[0017] M={(key target for chronic diseases 1, non-nutrient 1)k1

[0018] ,...,

[0019] (key targets for chronic diseases n, non-nutrient n) kn},

[0020] n≥500,

[0021] k is the chronic disease condition: chronic disease stage or chronic disease type,

[0022] Each key target of chronic disease includes examination data on oxidative stress, inflammation and metabolic disorder indicators;

[0023] Preprocessing the structured data set M: performing Z-score normalization on the continuous data under the same target index;

[0024] Dividing the structured data set M after the normalization process into a training set and a validation set;

[0025] Input the training set into a preset RF model, perform classification training on non-nutrients that intervene in the key targets of chronic diseases under different chronic disease conditions, and construct an initial chronic disease diet therapy recommendation model;

[0026] The classification performance of the initial chronic disease diet therapy recommendation model was verified using the validation set:

[0027] If the verification is passed, the chronic disease diet therapy recommendation model is deployed and applied on the chronic disease management server;

[0028] Otherwise, the structured dataset M is adjusted, the above steps are repeated, and the chronic disease diet therapy recommendation model is reconstructed.

[0029] Optionally, S3, the chronic disease status of the chronic disease patient and the examination data under each chronic disease assessment indicator are imported into the chronic disease diet therapy recommendation model, and the chronic disease diet therapy recommendation model recommends a corresponding non-nutritional vegetarian therapy list, including:

[0030] Identify the chronic disease type of the chronic disease patient according to the chronic disease condition;

[0031] Extracting the examination data under various oxidative stress, inflammation and metabolic disorder indicators;

[0032] Input the chronic disease type of the chronic disease patient and the examination data under various oxidative stress, inflammation and metabolic disorder indicators into the chronic disease diet therapy recommendation model, make a classification decision, and output the top three non-nutrients;

[0033] The chronic disease type of the chronic disease patient and the examination data under various oxidative stress, inflammation and metabolic disorder indicators and the output non-nutrients are written into the preset diet therapy list to generate the non-nutritional vegetarian diet therapy list and save it in the chronic disease electronic medical record file of the HIS system.

[0034] Optionally, the non-nutritional vegetarian therapy list also records the intake dosage D of the corresponding non-nutrients:

[0035]

[0036] HbA1C is an indicator of metabolic disorders.

[0037] D base For the basic dose of non-nutrients,

[0038] BMI refers to the BMI value of patients with chronic diseases.

[0039] Optionally, S4, extracting keywords from the chronic disease condition and the examination data under each chronic disease assessment indicator, and combining them with pre-built diet recommendation logic to generate corresponding prompt words and input them into a preset LLM large language model, using the LLM large language model to search the chronic disease recipe intervention database, and retrieve a recommended diet therapy list that matches the prompt words, including:

[0040] Extracting stage keywords related to the onset stage of the chronic disease from the chronic disease condition;

[0041] Extracting examination data keywords under various chronic disease assessment indicators from the examination data, including examination data feature values ​​under oxidative stress, inflammation, and metabolic disorder indicators;

[0042] Configuring a diet therapy recommendation logic that satisfies the stage keywords and the examination data keywords;

[0043] Combining the keywords and the diet therapy recommendation logic, constructing prompt words and inputting them into the LLM large language model preset on the chronic disease management server;

[0044] The LLM large language model is used to search the chronic disease recipe intervention database pre-deployed on the chronic disease management server, and the nutrients and non-nutrients matching the prompt words are retrieved and written into a preset structured list to obtain the recommended diet therapy list.

[0045] Optionally, S5, optimizing the recommended diet therapy list using the non-nutritional vegetarian diet therapy list, replacing or supplementing the non-nutrients therein, obtaining a complete diet therapy recipe for chronic diseases and recommending it to the user terminal, including:

[0046] Traversing the non-nutrients in the non-nutritional vegetarian therapy list and the recommended diet therapy list, identifying and marking the non-nutrients in the non-nutritional vegetarian therapy list that are not recorded in the recommended diet therapy list;

[0047] Replace or complete the marked non-nutrients in the recommended diet therapy list to obtain a complete diet therapy recipe for chronic diseases and save it in the chronic disease electronic medical record file in the HIS system;

[0048] The complete recipes for chronic disease diet therapy in the chronic disease electronic medical record file are sent to the bound user terminal on a regular basis through the chronic disease management server.

[0049] On the other hand, a device for intelligent optimization of chronic disease diet therapy by non-nutrient regulation of multiple key targets is provided, and the device for intelligent optimization of chronic disease diet therapy by non-nutrient regulation of multiple key targets is used to implement the above-mentioned method for intelligent optimization of chronic disease diet therapy by non-nutrient regulation of multiple key targets. The device includes a chronic disease management server and a user terminal, wherein:

[0050] The chronic disease management server is deployed with:

[0051] A chronic disease diet therapy recommendation model is used to recommend non-nutrients for chronic disease based on key targets of chronic diseases;

[0052] The HIS system is used to build electronic medical records for chronic disease patients and record their chronic disease management data, including real-time acquisition of chronic disease status and examination data under various chronic disease assessment indicators. The chronic disease assessment indicators include morphological indicators and physiological and biochemical indicators. The morphological indicators include waist circumference, blood pressure, height, weight or BMI. The physiological and biochemical indicators include the following key target indicators of chronic diseases: oxidative stress, inflammation and metabolic disorders.

[0053] A non-nutrient recommendation system is used to import the chronic disease status of chronic disease patients and the examination data under various chronic disease assessment indicators into the chronic disease diet therapy recommendation model, and the chronic disease diet therapy recommendation model recommends a corresponding non-nutrient vegetarian diet therapy list;

[0054] The chronic disease diet recommendation system is used to extract keywords from the examination data under the chronic disease status and each chronic disease assessment indicator, and combine it with the pre-built diet recommendation logic to generate corresponding prompt words and input them into the preset LLM large language model. The LLM large language model is used to search the chronic disease recipe intervention database to retrieve a recommended diet list that matches the prompt words;

[0055] A chronic disease recipe generation system is used to optimize the recommended diet therapy list using the non-nutritional vegetarian diet therapy list, replace or supplement the non-nutrients therein, and obtain a complete diet therapy recipe for chronic diseases;

[0056] A message recommendation system for recommending the excellent recipes for chronic disease diet therapy to the user terminal;

[0057] The chronic disease management server and the user terminal are communicatively connected.

[0058] On the other hand, an electronic device is provided, comprising: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, any one of the above-mentioned intelligent optimization methods for chronic disease diet therapy by non-nutrient regulation of multiple key targets is implemented.

[0059] On the other hand, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement any one of the above-mentioned intelligent optimization methods for chronic disease diet therapy by non-nutrient regulation of multiple key targets.

[0060] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0061] The present invention combines the non-nutrients that are affected by the key targets of chronic diseases to influence key targets, and recommends optimized dietary prescriptions based on non-nutrients to chronic disease patients with corresponding symptoms, thereby improving the antioxidant, anti-inflammatory, and metabolic adjustment of chronic disease patients. Through a dietary pattern based on the family nurse diet therapy theory, the impact of intervention on the metabolic indicators of patients with metabolic syndrome, a common basis for chronic diseases, is optimized, and a dietary pattern that can be promoted is formed. Combined with an intelligent algorithm model for optimized chronic disease diet therapy intervention, based on the basic principles of non-nutrient chronic disease intervention, it fills the gap in the unclear principles of healthy dietary patterns and related guidelines for disease treatment, and can track and manage the diet therapy of chronic disease patients through the system, dynamically adjust the non-nutritional vegetarian therapy list corresponding to the patient's real-time condition, and optimize the full-quality chronic disease diet therapy recipes, thereby promoting the treatment and rehabilitation of patients with chronic diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0063] Figure 1 This is a flow chart of an intelligent optimization method for chronic disease diet therapy by non-nutrient regulation of multiple key targets provided by an embodiment of the present invention;

[0064] Figure 2 This is a schematic structural diagram of a diet therapy theory model for family nurses provided by an embodiment of the present invention;

[0065] Figure 3 This is a block diagram of an intelligent optimization device for chronic disease diet therapy that uses multiple key targets for non-nutrient regulation, as provided by an embodiment of the present invention;

[0066] Figure 4 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0067] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0068] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0069] In the embodiments of the present invention, the terms "image" and "picture" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same. The terms "of," "corresponding," and "corresponding" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same.

[0070] In the embodiments of the present invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0071] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0072] The embodiment of the present invention provides an intelligent optimization method for chronic disease diet therapy by regulating non-nutrients through multiple key targets. The method can be implemented by an electronic device, which can be a terminal or a server. Figure 1 The flowchart of the intelligent optimization method for chronic disease diet therapy by non-nutrient regulation of multiple key targets is shown. The processing flow of the method may include the following steps:

[0073] S1. Based on the classification algorithm model, build and deploy a chronic disease diet recommendation model for non-nutrients based on key chronic disease targets and corresponding chronic disease conditions;

[0074] S2. Obtaining in real time the chronic disease status of chronic disease patients and examination data under various chronic disease assessment indicators from the chronic disease electronic medical records in the HIS system, wherein the chronic disease assessment indicators include morphological indicators and physiological and biochemical indicators. The morphological indicators include waist circumference, blood pressure, height, weight or BMI. The physiological and biochemical indicators include the following key target indicators of chronic diseases: oxidative stress, inflammation and metabolic disorders;

[0075] S3. Importing the chronic disease status of the chronic disease patient and the examination data under various chronic disease assessment indicators into the chronic disease diet therapy recommendation model, and having the chronic disease diet therapy recommendation model recommend a corresponding non-nutritional vegetarian diet therapy list;

[0076] S4. Extract keywords from the examination data under the chronic disease status and each chronic disease assessment indicator, and combine them with the pre-built diet recommendation logic to generate corresponding prompt words and input them into the preset LLM large language model. Use the LLM large language model to search the chronic disease recipe intervention database to retrieve a recommended diet therapy list that matches the prompt words;

[0077] S5. Optimize the recommended dietary therapy list using the non-nutritional vegetarian therapy list, replace or supplement the non-nutrients therein, obtain a complete and optimal diet recipe for chronic diseases, and recommend it to the user terminal;

[0078] S6. Update the electronic medical record of the chronic disease according to the diet therapy cycle, and repeat the above steps S2-S5.

[0079] This solution is implemented based on the chronic disease management server (referred to as the server).

[0080] like Figure 2 As shown, first describe the relationship between the key targets of chronic diseases (key targets affecting chronic diseases) and non-nutrients, as well as the family nurse diet therapy model (including family nurse diet therapy theory, family nurse diet therapy prescriptions and nurse diet therapy products):

[0081] In 2023, Han Shifan and others in my country first proposed the family nurse diet therapy theory. This theory explains the basic principles of non-nutrient chronic disease intervention from the perspective of the pathophysiology of chronic diseases, fills the gap in the healthy diet pattern and related guidelines for the treatment of diseases, and proposes a systematic modern diet therapy theoretical model to guide the diet therapy of chronic diseases. This study named the theory "family nurse diet therapy theory". And innovatively proposed a family nurse diet therapy theoretical model. This theoretical model takes the overall intervention of non-nutrients as the top, and regards the three elements of oxidative stress, inflammation and metabolic disorders that cause chronic diseases as an equilateral triangle regular triangular pyramid structure (see Figure 2 This model permeates the entire chronic disease management process, assisting in the treatment of chronic diseases by intervening at three key targets: oxidative stress, inflammation, and metabolic disorders. The application of non-nutrient dietary therapy is closely linked to the development of modern nutrition science, laying a solid foundation for the development of dietary therapy for family nurses.

[0082] Based on the family nurse diet therapy theory model, a non-nutrient chronic disease diet therapy program, also known as the family nurse diet therapy prescription (clinical registration number: ChiCTR2200066645), was proposed in 2022. This program observes and optimizes non-nutrient dietary patterns to improve the metabolic index levels of overweight / obese chronic disease patients, aiming to verify the effectiveness of the theory. Principles of diet therapy prescriptions guided by family nurse diet therapy theory:

[0083] In order to conform to the actual situation of residents' daily meal preparation and improve the feasibility and implementability of dietary intervention practices, precise measurement of meals is not required. Instead, reference values ​​of food types and weights are provided to try to conform to the family's daily dietary status.

[0084] The healthy dietary pattern selects food based on actual conditions under the guidance of family nurse diet therapy theory: fresh fruits and vegetables rich in non-nutrients are recommended, with a daily vegetable intake of more than 500g, 30-50g / d dietary fiber, ≥30% of which is soluble fiber, and whole grains are recommended as staple foods, accounting for at least one-third of daily staple food intake; in terms of fat intake, trans fatty acids are avoided, and it is recommended to replace saturated fatty acids with polyunsaturated fatty acids and monounsaturated fatty acids, such as edible olive oil, nuts and rapeseed oil, and the daily intake of nuts as snacks between meals should not exceed 50g; high-quality protein sources are recommended for daily protein, such as meat, eggs, beans, and milk, and fish, chicken breast, lean beef, lamb, etc. are recommended for meat; patients are recommended to choose foods with a low glycemic index to facilitate blood sugar control.

[0085] Calculate daily calorie requirements: Calculate daily energy requirements based on the patient's BMI and physical activity level. Total daily calorie requirement = standard body weight x daily calorie intake per kilogram of body weight. A daily diet should be developed for the patient based on the food exchange method. Daily energy proportions: 1 / 3 of the daily energy intake should be consumed for breakfast, lunch, and dinner, with two snacks between the three main meals. Snack foods should be 10% of the energy intake from the main meals. Follow the principle of eating small, frequent meals.

[0086] Therefore, the present invention conducted research on the effects of optimized non-nutrient dietary prescriptions on the antioxidant, anti-inflammatory, and metabolic regulation of patients with chronic diseases. The research mainly aimed to optimize the actual effects of previous dietary therapy plans and form a dietary pattern that can be promoted by intervening in dietary patterns based on family nurse dietary therapy theory on the metabolic indicators of patients with metabolic syndrome, a common basis for chronic diseases.

[0087] And nurse dietetic products, such as:

[0088] Substitute tea is a brewed tea made with edible buckwheat grains. This product complies with GH / T 1091-2014 and has a production license number of SC11461012208320.

[0089] Protein Compound Powder is a meal replacement product with an optimized non-nutrient ratio. It uses soy protein isolate as its primary ingredient and contains key non-nutrient ingredients such as flavonoids, polysaccharides, sterols, β-glucan, and alkaloids. This product complies with the implementation standards of GB / T 29602 and has a production license number of SC106210213011970. It's important to note that our product has completed pilot testing.

[0090] Specifically, the present invention integrates multi-source heterogeneous data of patients with chronic diseases (including biochemical indicators, morphological indicators, disease stages, and gene expression data), and extracts non-nutrients (such as curcumin and resveratrol) that are highly correlated with chronic disease targets (such as the inflammatory factor IL-6 and the oxidative stress indicator MDA) through feature engineering.

[0091] Using a random forest (RF) or XGBoost model, the input is a standardized multimodal feature vector, and the output is a probability distribution of non-nutrient recommendations. A chronic disease dietary recommendation model is constructed to recommend non-nutrients based on key chronic disease targets, enabling intelligent non-nutrient recommendations.

[0092] Optionally, S1. Based on the classification algorithm model, a chronic disease diet recommendation model is constructed and deployed for recommending non-nutrients for corresponding chronic disease conditions based on key targets of chronic diseases, including:

[0093] Collect a structured dataset M consisting of non-nutrients that intervene in the key targets of chronic diseases under different chronic disease conditions and the key targets of chronic diseases:

[0094] M={(key target for chronic diseases 1, non-nutrient 1)k1

[0095] ,...,

[0096] (key targets for chronic diseases n, non-nutrient n) kn},

[0097] n≥500,

[0098] k is the chronic disease condition: chronic disease stage or chronic disease type,

[0099] Each key target of chronic disease includes examination data on oxidative stress, inflammation and metabolic disorder indicators;

[0100] Preprocessing the structured data set M: performing Z-score normalization on the continuous data under the same target index;

[0101] Dividing the structured data set M after the normalization process into a training set and a validation set;

[0102] Input the training set into a preset RF model, perform classification training on non-nutrients that intervene in the key targets of chronic diseases under different chronic disease conditions, and construct an initial chronic disease diet therapy recommendation model;

[0103] The classification performance of the initial chronic disease diet therapy recommendation model was verified using the validation set:

[0104] If the verification is passed, the chronic disease diet therapy recommendation model is deployed and applied on the chronic disease management server;

[0105] Otherwise, the structured dataset M is adjusted, the above steps are repeated, and the chronic disease diet therapy recommendation model is reconstructed.

[0106] First, chronic disease assessment indicators include morphological indicators and physiological and biochemical indicators. The morphological indicators include waist circumference, blood pressure, height, weight, or BMI. The physiological and biochemical indicators include the following key chronic disease target indicators: oxidative stress, inflammation, and metabolic disorders (such as fasting plasma glucose (FPG), fasting insulin (FINS), glycosylated hemoglobin (HbA1c), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C), blood urea nitrogen (BUN), uric acid (UA), plasma homocysteine ​​(Hcy), and insulin resistance index (HOMA-IR)). The key chronic disease targets here refer to the examination data under oxidative stress, inflammation, and metabolic disorders.

[0107] Secondly, a structured data set M consisting of non-nutrients that intervene in the key targets of chronic diseases under different chronic disease conditions and the key targets of chronic diseases can be collected:

[0108] M={(key target for chronic diseases 1, non-nutrient 1)k1

[0109] ,...,

[0110] (key targets for chronic diseases n, non-nutrient n) kn},

[0111] n≥500,

[0112] k is the chronic disease condition: chronic disease stage or chronic disease type,

[0113] Each key target of chronic disease includes examination data on oxidative stress, inflammation and metabolic disorder indicators;

[0114] Preprocessing the structured data set M: performing Z-score normalization on the continuous data under the same target index;

[0115] For example, patients with stage II diabetes may have examination data on oxidative stress, inflammation, and metabolic disorders, such as CRP > 5mg / L. These chronic disease indicator data for different chronic disease stages or types can be found in the database's historical records, so the dataset is obtained from the server's database.

[0116] In the past, doctors would recommend corresponding non-nutrients to patients for each group of chronic disease indicator data corresponding to different chronic disease stages or types. Therefore, corresponding data sets were recorded in the information of the database, thereby obtaining a structured data set M (stored in the MySQL database on the server) consisting of several groups of non-nutrients that intervene in the key targets of the chronic disease under different chronic disease conditions and data groups of the key targets of the chronic disease.

[0117] In order to facilitate the subsequent unified data processing, Z-score standardization is performed on continuous target indicators:

[0118]

[0119] X: original data;

[0120] μ: mean;

[0121] σ: standard deviation.

[0122] After obtaining the structured dataset M, divide it into a training set and a validation set;

[0123] The training set is input into the preset RF model, and non-nutrients that intervene in the key targets of chronic diseases under different chronic disease conditions are classified and trained to construct the initial chronic disease diet therapy recommendation model; wherein, the RF model is constructed as follows:

[0124] Input: standardized target index + chronic disease type;

[0125] Output: non-nutrient recommendation probability distribution;

[0126] Parameters: number of trees = 100, maximum depth = 10, splitting criterion = Gini coefficient.

[0127] The general training and validation of the RF model will not be described here in detail. You can combine the general training steps of the RF model with the performance validation methods, such as F1 or AUC curves.

[0128] Technical advantages:

[0129] Anti-overfitting: Random forest reduces variance through bagging and random feature selection.

[0130] Interpretability: Output feature importance ranking (such as the weight of inflammation indicators > 60%).

[0131] Before implementing non-nutrient-based chronic disease diet therapy, first obtain the chronic disease status of chronic disease patients and examination data under various chronic disease assessment indicators in real time from the chronic disease electronic medical records of the HIS system. The chronic disease assessment indicators include morphological indicators and physiological and biochemical indicators. The morphological indicators include waist circumference, blood pressure, height, weight or BMI. The physiological and biochemical indicators include the following chronic disease key target indicators: oxidative stress, inflammation and metabolic disorders. The main evaluation indicators include:

[0132] Morphological indicators, such as waist circumference, blood pressure, height, weight, and BMI;

[0133] Physiological and biochemical indicators: oxidative stress, inflammation and metabolic disorders, such as fasting blood glucose (FPG), fasting insulin (FINS), glycosylated hemoglobin (HbA1c), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C), blood urea nitrogen (BUN), uric acid (UA), plasma homocysteine ​​(Hcy), and insulin resistance index (HOMA-IR). These data can be recorded by the hospital system, and the patient's various indicator examination data can be uploaded to the backend server by the hospital examination and stored in the patient's chronic disease electronic medical record file. The Apache Kafka stream processing engine can be used to dynamically clean the data (remove outliers) and store it in a NoSQL database (MongoDB).

[0134] Dietary therapy prescription process under the guidance of family nurse diet therapy theory:

[0135] The case manager will evaluate the patient's disease condition and examination indicators, and build the patient's diet therapy file based on the evaluation results.

[0136] After obtaining the disease status information and examination data of various indicators of chronic disease patients, we can start to recommend recipes for them.

[0137] Optionally, S3, the chronic disease status of the chronic disease patient and the examination data under each chronic disease assessment indicator are imported into the chronic disease diet therapy recommendation model, and the chronic disease diet therapy recommendation model recommends a corresponding non-nutritional vegetarian therapy list, including:

[0138] Identify the chronic disease type of the chronic disease patient according to the chronic disease condition;

[0139] Extracting the examination data under various oxidative stress, inflammation and metabolic disorder indicators;

[0140] Input the chronic disease type of the chronic disease patient and the examination data under various oxidative stress, inflammation and metabolic disorder indicators into the chronic disease diet therapy recommendation model, make a classification decision, and output the top three non-nutrients;

[0141] The chronic disease type of the chronic disease patient and the examination data under various oxidative stress, inflammation and metabolic disorder indicators and the output non-nutrients are written into the preset diet therapy list to generate the non-nutritional vegetarian diet therapy list and save it in the chronic disease electronic medical record file of the HIS system.

[0142] Because a classification model was previously constructed to recommend corresponding non-nutrients based on the chronic disease type of chronic disease patients and the examination data under various oxidative stress, inflammation and metabolic disorder indicators, the "Chronic Disease Dietary Therapy Recommendation Model" can classify and output the matching non-nutrients based on the chronic disease type of chronic disease patients and the examination data under various oxidative stress, inflammation and metabolic disorder indicators. Specifically, the non-nutrient vegetarian therapy list is generated:

[0143] Model Inference:

[0144] Input patient data into the RF model and output the top 3 non-nutrient recommendations (such as curcumin, quercetin, and EGCG);

[0145] Recommended basis: Probability threshold filtering (only output items with probability > 0.7).

[0146] For non-nutrients that are recommended, such as 500mg of curcumin, which matches the patient's condition and indicators, the dosage needs to be further determined. The dosage is calculated based on the patient's physical signs:

[0147] Optionally, the non-nutritional vegetarian therapy list also records the intake dosage D of the corresponding non-nutrients:

[0148]

[0149] HbA1C is an indicator of metabolic disorders.

[0150] D base For the basic dose of non-nutrients,

[0151] BMI refers to the BMI value of patients with chronic diseases.

[0152] By dynamically adjusting the dosage based on the patient's physical signs, we can avoid "one-size-fits-all" recommendations for non-nutrients and make targeted recommendations based on the patient's physical signs, which is more in line with the patient's chronic disease status.

[0153] Optionally, S4, extracting keywords from the chronic disease condition and the examination data under each chronic disease assessment indicator, and combining them with pre-built diet recommendation logic to generate corresponding prompt words and input them into a preset LLM large language model, using the LLM large language model to search the chronic disease recipe intervention database, and retrieve a recommended diet therapy list that matches the prompt words, including:

[0154] Extracting stage keywords related to the onset stage of the chronic disease from the chronic disease condition;

[0155] Extracting examination data keywords under various chronic disease assessment indicators from the examination data, including examination data feature values ​​under oxidative stress, inflammation, and metabolic disorder indicators;

[0156] Configuring a diet therapy recommendation logic that satisfies the stage keywords and the examination data keywords;

[0157] Combining the keywords and the diet therapy recommendation logic, constructing prompt words and inputting them into the LLM large language model preset on the chronic disease management server;

[0158] The LLM large language model is used to search the chronic disease recipe intervention database pre-deployed on the chronic disease management server, and the nutrients and non-nutrients matching the prompt words are retrieved and written into a preset structured list to obtain the recommended diet therapy list.

[0159] First, a chronic disease recipe intervention database is deployed on the server. It is constructed based on "nutrients, non-nutrients, diseases, indicators, mechanisms", etc. Based on nutrients, non-nutrients and the related structures and matching applications between the two, and using big data technology, nutrients and non-nutrients that intervene in the development of chronic diseases are constructed. That is, a structured data set of nutrients or non-nutrients and their dosages for intervening in chronic diseases under various indicators is stored. Specifically, the database should include common types of foods and medicinal and edible substances; the content, efficacy (anti-inflammatory, antioxidant, improving metabolism) of non-nutrients in common foods, improvement of disease indicators, and targets of action; a summary of literature evidence on the prevention and treatment of chronic diseases with non-nutrients (graded: clinical trials, animal experiments, cell experiments), etc. The database can be composed of a food storage module (which records the IDs of different foods and can perform data operations and permission settings), a disease location mapping module (which can establish non-nutritional substances associated with different diseases, and the administrator can add, delete and query in this module), a target module (which locates the disease targets caused by various diseases such as inflammation, oxidative stress, etc.), a signal pathway management module (used to manage target gene signals associated with non-nutritional substances and various targets) and a non-nutritional substance recording module (which records several types of non-nutritional substances, uses the structure of the MeSH database for structured storage, and records the molecular formula, structural formula and other information of the non-nutritional substances for easy query).

[0160] Secondly, we recommend to patients a universal diet that matches their chronic disease status and the examination data under various chronic disease assessment indicators, that is, a recommended diet list that includes nutrients and non-nutrients (however, the non-nutrients here are not recommended separately, so the output of the classification model will be used for optimization in the future to optimize and improve the patient's non-nutrients).

[0161] During the specific implementation, we used the LLM large language model to assist in generating the recommended dietary therapy list:

[0162] First, we use text recognition algorithms to extract key words related to the stages of chronic disease onset and key words for examination data under various chronic disease assessment indicators, including characteristic values ​​of examination data under oxidative stress, inflammation, and metabolic disorders. For example:

[0163] “Diabetes stage II + CRP>5mg / L + BMI=28”.

[0164] When building the diet therapy recommendation logic, it is necessary to retrieve the recommended diet therapy list that meets the stage keywords and the examination data keywords. At this time, you can combine them to build prompt words, such as:

[0165] Template: "Recommend a diet that contains [non-nutrients] and meets [BMI=XX][CRP=XX] for patients with [chronic disease type]."

[0166] Example: "Recommend a diet containing quercetin suitable for patients with stage II diabetes mellitus and a BMI of 28 and CRP of 6 mg / L."

[0167] The LLM large language model can be developed by the user and deployed on the "chronic disease management server", or an API port that calls a third-party LLM large language model can be deployed on the chronic disease management server.

[0168] LLM Search and Matching:

[0169] The prompt words are encoded into vectors using the BERT model. The chronic disease recipe database is retrieved through cosine similarity (the algorithm is not described in detail), and the top 5 similar recipes (such as "bitter melon scrambled eggs + quercetin supplement") are returned. These retrieved nutrients and non-nutrients are written into a preset structured list to obtain the recommended diet therapy list.

[0170] The BERT model captures implicit requirements such as "low GI" and "anti-inflammatory." It supports multi-condition matching across text, nutrition information, and images.

[0171] Therefore, using LLM search can quickly obtain a universal recommended dietary therapy list for patients.

[0172] Optionally, S5, optimizing the recommended diet therapy list using the non-nutritional vegetarian diet therapy list, replacing or supplementing the non-nutrients therein, obtaining a complete diet therapy recipe for chronic diseases and recommending it to the user terminal, including:

[0173] Traversing the non-nutrients in the non-nutritional vegetarian therapy list and the recommended diet therapy list, identifying and marking the non-nutrients in the non-nutritional vegetarian therapy list that are not recorded in the recommended diet therapy list;

[0174] Replace or complete the marked non-nutrients in the recommended diet therapy list to obtain a complete diet therapy recipe for chronic diseases and save it in the chronic disease electronic medical record file in the HIS system;

[0175] The complete recipes for chronic disease diet therapy in the chronic disease electronic medical record file are sent to the bound user terminal on a regular basis through the chronic disease management server.

[0176] Here we use a differential marking algorithm to traverse and identify differential non-nutrients, specifically:

[0177] Use a hash table to quickly compare the non-nutrient names in two lists:

[0178]

[0179] Output: Flag non-nutrients (e.g., “curcumin”) that are present in the RF list but missing from the LLM list.

[0180] Efficient comparison is achieved through hash tables (traditional loop comparison requires O(n 2 )); Mark the priority of missing items according to the recommendation probability (priority field) of the RF model to achieve dynamic priority sorting.

[0181] Non-nutrient replacement and completion, replacement logic: If there are low-priority non-nutrients in the LLM list (such as priority < 0.7), replace them with high-priority items in the RF list:

[0182]

[0183]

[0184] Completion logic: If no replacement is required, add the missing item directly (marking the source as "supplement" or specific food).

[0185] Technical advantages: Prioritize replacement of low-correlation ingredients (e.g. EGCG priority 0.65 is replaced by curcumin 0.85); clarify non-nutrient sources (e.g. "Curcumin: supplement") to improve patient compliance.

[0186] Storage and push of excellent recipes for chronic disease diet therapy

[0187] HIS system integration: Write the full recipe to the patient's electronic medical record NutritionOrder resource through the FHIR standard API:

[0188]

[0189] Scheduled push logic: Use Apache Kafka to set up a scheduled task (Cron Job) to trigger push based on the patient's follow-up period:

[0190] #Push the daily recipe at 8:00 every day

[0191] 0 8***curl-X POST-H"Content-Type:application / json"-d'{"patient_id":"P001"}'http: / / server / api / push_recipe.

[0192] Push content: including text recipes, graphic descriptions and links to purchase supplements (such as hospital-partnered pharmacies).

[0193] Supports setting the push frequency according to the patient's personalized needs (such as daily push for diabetic patients, weekly push for hypertension patients). The user end can be an APP terminal.

[0194] Apply the above method for clinical verification:

[0195] A randomized, single-blind design was used to recruit and complete an intervention study of 70 patients with metabolic syndrome. Researchers provided fully informed consent to participants who met the inclusion criteria. The intervention group, consisting of an optimized non-nutrient diet group, implemented a meal replacement diet model based on the dietary therapy theory of family nurses. Based on the specific characteristics of each participant, a personalized diet menu based on non-nutrients was constructed. The staple food on the menu was family nurse meal replacement powder, with one bag (20g) for each breakfast, lunch, and dinner, mixed with 300mL of warm water. The rest of the menu was the same as for the non-nutrient-rich diet group. The participants' original medication and exercise status remained unchanged. The participants were informed in detail of the specific content and precautions of the diet to improve their compliance and clinical effectiveness. The control group received a non-nutrient-rich diet. A personalized diet plan focused on non-nutrients was developed based on the subjects' weight, dietary habits, occupation, and medical conditions. The staple food on the plan consisted of whole grains, and the remaining foods were high in non-nutrients. Family nurse substitute tea (primarily composed of buckwheat and wolfberry) was used as daily drinking water to increase non-nutrient content. Each bag contained 3g of tea, two bags per day, brewed with 300mL of hot water at or above 80°C for at least 3 minutes. Drink the tea at least three times daily to maximize its effectiveness. The plan consisted of a one-week regimen (consisting of a fixed amount of staple food, fruits and vegetables, and lean meat, with appropriate amounts of nuts and dairy products added between meals). Dietary patterns were adjusted promptly based on subject feedback.

[0196] Outcome measures: 1) Primary outcome measures, namely body weight and insulin resistance index; 2) Secondary outcome measures, including anthropometric indicators (BMI, waist circumference), biochemical indicators (glucose metabolism indicators (FPG, FINS), lipid metabolism indicators (TC, TG, HDL-C, LDL-C), liver and kidney function indicators (alanine aminotransferase, aspartate aminotransferase, blood urea nitrogen, blood creatinine, uric acid), homocysteine), inflammatory indicators (high-sensitivity-C reactive protein), and oxidative stress indicators (malondialdehyde, superoxide dismutase). All indicators were measured before and 2 days after the intervention. The results showed that both dietary patterns could reduce the subjects' weight and had certain effects on improving glucose and lipid metabolism indicators and liver and kidney function. The optimized non-nutrient diet group had a significantly better improvement in weight, FINS, and uric acid indicators than the diet rich in non-nutrients. At the same time, the optimized non-nutrient diet pattern could reduce the subjects' systolic blood pressure and homocysteine, improve inflammatory indicators, and increase superoxide dismutase activity. Therefore, after rigorous clinical verification, the research program confirmed the effects of the family nurse diet therapy research program in anti-inflammation, anti-oxidation and metabolic adjustment, forming a replicable dietary prescription model.

[0197] Therefore, this solution has the following advantages:

[0198] 1. Fill the gaps in medical principles:

[0199] Based on the molecular mechanism of non-nutrient intervention in chronic diseases (such as curcumin inhibiting COX-2 enzyme), the biomedical basis of the dietary therapy plan is clarified, breaking through the vague advice of "eating more vegetables and fruits" in traditional guidelines.

[0200] 2. Dynamic personalized intervention:

[0201] Real-time integration of multimodal patient data (biochemical indicators + behavioral data) allows for dynamic adjustment of dietary therapy plans (e.g., increasing the recommended weight of Omega-3 when inflammation worsens).

[0202] 3. Integration of intelligence and tradition:

[0203] Encode family nurse diet therapy theories (such as phased nutritional intervention) into a rule engine and collaborate with the AI ​​model to ensure that the plan is both scientific and practical (such as automatically filtering solid recipes during the postoperative liquid diet stage).

[0204] 4. Full-process closed-loop management:

[0205] A closed loop is formed from data collection → model recommendation → LLM retrieval → optimization feedback, and the efficacy rate is improved by 50% (compared with traditional manual nutrition plans).

[0206] 5. Evidence-based medicine verification:

[0207] Verified by RCT clinical trials, after 6 months of use, the improvement rate of metabolic syndrome indicators (waist circumference, fasting blood sugar) in overweight / obese patients was >40%, and the antioxidant index (SOD activity) increased by 25%.

[0208] Figure 3 This is a block diagram of an intelligent optimization device for chronic disease diet therapy that uses multiple key targets for non-nutrient regulation, according to an exemplary embodiment. The device is used in an intelligent optimization method for chronic disease diet therapy that uses multiple key targets for non-nutrient regulation. On the other hand, a device for intelligent optimization of chronic disease diet therapy that uses multiple key targets for non-nutrient regulation is provided. The device is used to implement the above-mentioned intelligent optimization method for chronic disease diet therapy that uses multiple key targets for non-nutrient regulation. The device includes a chronic disease management server and a user terminal, wherein:

[0209] The chronic disease management server is deployed with:

[0210] A chronic disease diet therapy recommendation model is used to recommend non-nutrients for chronic disease based on key targets of chronic diseases;

[0211] The HIS system is used to build electronic medical records for chronic disease patients and record their chronic disease management data, including real-time acquisition of chronic disease status and examination data under various chronic disease assessment indicators. The chronic disease assessment indicators include morphological indicators and physiological and biochemical indicators. The morphological indicators include waist circumference, blood pressure, height, weight or BMI. The physiological and biochemical indicators include the following key target indicators of chronic diseases: oxidative stress, inflammation and metabolic disorders.

[0212] A non-nutrient recommendation system is used to import the chronic disease status of chronic disease patients and the examination data under various chronic disease assessment indicators into the chronic disease diet therapy recommendation model, and the chronic disease diet therapy recommendation model recommends a corresponding non-nutrient vegetarian diet therapy list;

[0213] The chronic disease diet recommendation system is used to extract keywords from the examination data under the chronic disease status and each chronic disease assessment indicator, and combine it with the pre-built diet recommendation logic to generate corresponding prompt words and input them into the preset LLM large language model. The LLM large language model is used to search the chronic disease recipe intervention database to retrieve a recommended diet list that matches the prompt words;

[0214] A chronic disease recipe generation system is used to optimize the recommended diet therapy list using the non-nutritional vegetarian diet therapy list, replace or supplement the non-nutrients therein, and obtain a complete diet therapy recipe for chronic diseases;

[0215] A message recommendation system for recommending the excellent recipes for chronic disease diet therapy to the user terminal;

[0216] The chronic disease management server and the user terminal are communicatively connected.

[0217] The interaction and composition of the above devices should be understood in conjunction with the above method steps, and will not be described in detail here.

[0218] Figure 4 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention, such as Figure 4 As shown, the electronic device 410 may include a first processor 2001 .

[0219] Optionally, the electronic device 410 may further include a memory 2002 and a transceiver 2003 .

[0220] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0221] The following combination Figure 4 The components of the electronic device 410 are described in detail.

[0222] The first processor 2001 is the control center of the electronic device 410 and can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement an embodiment of the present invention, such as one or more digital signal processors (DSPs) or one or more field programmable gate arrays (FPGAs).

[0223] Optionally, the first processor 2001 can execute various functions of the electronic device 410 by running or executing software programs stored in the memory 2002 and calling data stored in the memory 2002.

[0224] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 4 CPU0 and CPU1 are shown in FIG.

[0225] In a specific implementation, as an embodiment, the electronic device 410 may also include multiple processors, such as Figure 4 1 and 2. The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). A processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0226] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled by the first processor 2001 for execution. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0227] Alternatively, the memory 2002 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently and be accessed through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0228] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0229] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 4 (not shown separately in the figure). The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0230] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0231] It should be noted that Figure 4 The structure of the electronic device 410 shown in the figure does not constitute a limitation on the router. The actual knowledge structure recognition device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0232] In addition, the technical effects of the electronic device 410 can refer to the technical effects of the intelligent optimization method for chronic disease diet therapy by non-nutrient regulation of multiple key targets described in the above method embodiment, and will not be repeated here.

[0233] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0234] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0235] The above embodiments can be implemented in whole or in part through software, hardware (such as circuits), firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired method (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, or magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0236] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0237] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0238] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0239] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0240] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0241] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of the device or unit, which can be electrical, mechanical or other forms.

[0242] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0243] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0244] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0245] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. An intelligent optimization method for chronic disease diet therapy by regulating non-nutrients through multiple key targets, characterized in that: The method comprises: S1. Based on the classification algorithm model, build and deploy a chronic disease diet recommendation model for non-nutrients based on key chronic disease targets and corresponding chronic disease conditions; S2. Obtaining in real time the chronic disease status of chronic disease patients and examination data under various chronic disease assessment indicators from the chronic disease electronic medical records in the HIS system, wherein the chronic disease assessment indicators include morphological indicators and physiological and biochemical indicators. The morphological indicators include waist circumference, blood pressure, height, weight or BMI. The physiological and biochemical indicators include the following key target indicators of chronic diseases: oxidative stress, inflammation and metabolic disorders; S3. Importing the chronic disease status of the chronic disease patient and the examination data under various chronic disease assessment indicators into the chronic disease diet therapy recommendation model, and having the chronic disease diet therapy recommendation model recommend a corresponding non-nutritional vegetarian diet therapy list; S4. Extract keywords from the examination data under the chronic disease status and each chronic disease assessment indicator, and combine them with the pre-built diet recommendation logic to generate corresponding prompt words and input them into the preset LLM large language model. Use the LLM large language model to search the chronic disease recipe intervention database to retrieve a recommended diet therapy list that matches the prompt words; S5. Optimize the recommended dietary therapy list using the non-nutritional vegetarian therapy list, replace or supplement the non-nutrients therein, obtain a complete and optimal diet recipe for chronic diseases, and recommend it to the user terminal; S6. Update the electronic medical record of the chronic disease according to the diet therapy cycle, and repeat the above steps S2-S5.

2. The intelligent optimization method for chronic disease diet therapy by regulating non-nutrients through multiple key targets according to claim 1, characterized in that: S1. Based on the classification algorithm model, build and deploy a chronic disease diet recommendation model for non-nutrients based on key chronic disease targets and corresponding chronic disease conditions, including: Collect a structured dataset M consisting of non-nutrients that intervene in the key targets of chronic diseases under different chronic disease conditions and the key targets of chronic diseases: M={(key target for chronic diseases 1, non-nutrient 1) k1 ，...， (Key targets for chronic diseases n, non-nutrient n) kn }, n≥500， k is the chronic disease condition: chronic disease stage or chronic disease type, Each key target of chronic disease includes examination data on oxidative stress, inflammation and metabolic disorder indicators; Preprocessing the structured data set M: performing Z-score normalization on the continuous data under the same target index; Dividing the structured data set M after the normalization process into a training set and a validation set; Input the training set into a preset RF model, perform classification training on non-nutrients that intervene in the key targets of chronic diseases under different chronic disease conditions, and construct an initial chronic disease diet therapy recommendation model; The classification performance of the initial chronic disease diet therapy recommendation model was verified using the validation set: If the verification is passed, the chronic disease diet therapy recommendation model is deployed and applied on the chronic disease management server; Otherwise, the structured dataset M is adjusted, the above steps are repeated, and the chronic disease diet therapy recommendation model is reconstructed.

3. The intelligent optimization method for chronic disease diet therapy by regulating non-nutrients through multiple key targets according to claim 2, characterized in that: S3. Import the chronic disease status of the chronic disease patient and the examination data under various chronic disease assessment indicators into the chronic disease diet therapy recommendation model, and the chronic disease diet therapy recommendation model recommends a corresponding non-nutritional vegetarian diet therapy list, including: Identify the chronic disease type of the chronic disease patient according to the chronic disease condition; Extracting the examination data under various oxidative stress, inflammation and metabolic disorder indicators; Input the chronic disease type of the chronic disease patient and the examination data under various oxidative stress, inflammation and metabolic disorder indicators into the chronic disease diet therapy recommendation model, make a classification decision, and output the top three non-nutrients; The chronic disease type of the chronic disease patient and the examination data under various oxidative stress, inflammation and metabolic disorder indicators and the output non-nutrients are written into the preset diet therapy list to generate the non-nutritional vegetarian diet therapy list and save it in the chronic disease electronic medical record file of the HIS system.

4. The intelligent optimization method for chronic disease diet therapy by regulating non-nutrients through multiple key targets according to claim 3, characterized in that: The non-nutritional vegetarian therapy list also records the corresponding non-nutrient intake dose D: HbA1C is an indicator of metabolic disorders. D base For the basic dose of non-nutrients, BMI refers to the BMI value of patients with chronic diseases.

5. The intelligent optimization method for chronic disease diet therapy by regulating non-nutrients through multiple key targets according to claim 2, characterized in that: S4. Extract keywords from the examination data under the chronic disease condition and each chronic disease assessment indicator, and combine them with the pre-built diet recommendation logic to generate corresponding prompt words and input them into the preset LLM large language model. Use the LLM large language model to search the chronic disease recipe intervention database and retrieve a recommended diet therapy list that matches the prompt words, including: Extracting stage keywords related to the onset stage of the chronic disease from the chronic disease condition; Extracting examination data keywords under various chronic disease assessment indicators from the examination data, including examination data feature values ​​under oxidative stress, inflammation, and metabolic disorder indicators; Configuring a diet therapy recommendation logic that satisfies the stage keywords and the examination data keywords; Combining the keywords and the diet therapy recommendation logic, constructing prompt words and inputting them into the LLM large language model preset on the chronic disease management server; The LLM large language model is used to search the chronic disease recipe intervention database pre-deployed on the chronic disease management server, and the nutrients and non-nutrients matching the prompt words are retrieved and written into a preset structured list to obtain the recommended diet therapy list.

6. The intelligent optimization method for chronic disease diet therapy by regulating non-nutrients through multiple key targets according to claim 4, characterized in that: S5. Optimizing the recommended dietary therapy list using the non-nutritional vegetarian therapy list, replacing or supplementing the non-nutrients therein, obtaining a complete and optimal diet therapy recipe for chronic diseases and recommending it to the user terminal, including: Traversing the non-nutrients in the non-nutritional vegetarian therapy list and the recommended diet therapy list, identifying and marking the non-nutrients in the non-nutritional vegetarian therapy list that are not recorded in the recommended diet therapy list; Replace or complete the marked non-nutrients in the recommended diet therapy list to obtain a complete diet therapy recipe for chronic diseases and save it in the chronic disease electronic medical record file in the HIS system; The complete recipes for chronic disease diet therapy in the chronic disease electronic medical record file are sent to the bound user terminal on a regular basis through the chronic disease management server.

7. An intelligent optimization device for chronic disease diet therapy by non-nutrient regulation by multiple key targets, the intelligent optimization device for chronic disease diet therapy by non-nutrient regulation by multiple key targets is used to implement the intelligent optimization method for chronic disease diet therapy by non-nutrient regulation by multiple key targets as described in any one of claims 1-6, characterized in that: The device includes a chronic disease management server and a user terminal, wherein: The chronic disease management server is deployed with: A chronic disease diet therapy recommendation model is used to recommend non-nutrients for chronic disease based on key targets of chronic diseases; The HIS system is used to build electronic medical records for chronic disease patients and record their chronic disease management data, including real-time acquisition of chronic disease status and examination data under various chronic disease assessment indicators. The chronic disease assessment indicators include morphological indicators and physiological and biochemical indicators. The morphological indicators include waist circumference, blood pressure, height, weight or BMI. The physiological and biochemical indicators include the following key target indicators of chronic diseases: oxidative stress, inflammation and metabolic disorders. A non-nutrient recommendation system is used to import the chronic disease status of chronic disease patients and the examination data under various chronic disease assessment indicators into the chronic disease diet therapy recommendation model, and the chronic disease diet therapy recommendation model recommends a corresponding non-nutrient vegetarian diet therapy list; The chronic disease diet recommendation system is used to extract keywords from the examination data under the chronic disease status and each chronic disease assessment indicator, and combine it with the pre-built diet recommendation logic to generate corresponding prompt words and input them into the preset LLM large language model. The LLM large language model is used to search the chronic disease recipe intervention database to retrieve a recommended diet list that matches the prompt words; A chronic disease recipe generation system is used to optimize the recommended diet therapy list using the non-nutritional vegetarian diet therapy list, replace or supplement the non-nutrients therein, and obtain a complete diet therapy recipe for chronic diseases; A message recommendation system for recommending the excellent recipes for chronic disease diet therapy to the user terminal; The chronic disease management server and the user terminal are communicatively connected.

8. An electronic device, characterized in that: The electronic device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program code, which can be called by a processor to execute the method according to any one of claims 1 to 6.

Images